Introduction

In a 2014 investigation, The Washington Post reported that the FBI had been under-counting fatal police shootings by more than half. An eighteen year-old Michael Brown was killed by police in St. Louis, Missouri, shining a light on the issue of police brutality in the United States. Relying on news, social media, and police reports, the Washington Post compiled a comprehensive list of police shootings from 2015 to 2022 to amend the issue of under-counting police killings of civilians. Police brutality stems from a lack of accountability. The police force is consistently protected from the consequences of violence, regardless of the use of such tools of accountability as police body cameras. We studied the use of body cameras in police shootings to understand whether the use of body cameras was more common in one state over another. More specifically, we asked: within our data set of 6,574 observations of police shootings from 2015 to 2022 in the United States, is there a correlation between the U.S. state of observation and whether a body camera was turned on during the shooting? We found that the use of body cameras during fatal police shootings certainly varies by state, pointing to the conclusion that state policy plays a role in whether body cameras are turned on during active police duty.

This data set includes information on every fatal police shooting from January 1, 2015 to the present. It is updated regularly by the Washington Post. Some of the Post’s key findings from the data set include:

  1. Overall, police nationwide kill approximately 1,000 civilians annually.

  2. Black people are killed by police at a higher rate than White people. In absolute numbers, the White population is killed most often by police. Black people, accounting for less than 13 percent of the population, are killed more than twice the rate of White people.

  3. The majority of victims are male and between the ages of 20 and 40 years.

  4. Police shootings occur in every state. In cities, they are more frequent. The states with the highest rates of shootings (considering state population) are New Mexico, Alaska, and Oklahoma.

These findings are the result of previous research by the Washington Post. One of the data set’s limitations is the absence of data on populations at the time of the observation. This information would be valuable for any type of racial injustice analysis. The research we gathered through the EDA began with very basic analyses:

These basic univariate analyses were created to better understand the data set and to identify any interesting patterns that emerged. We began doing more complex multivariate analyses and noting any interesting findings:

Hover over the map below to see the breakdown of fatal police shootings, divided by the race of the victim. We looked at the total number of deaths in each state by race and following are some of the insights:

  1. We see that the state with the highest level of victims of police violence is California with a total of 885 victims, followed by Texas with a total of 553 and then Florida with 427.

  2. These results are consistent with the populations of these states, with the highest being California, then Texas, and then Florida.

  3. We also observe that the highest number of deaths is for Hispanic people in California, whereas in Texas and Florida there are more fatal shootings of White people.

Then we looked at the age of the suspect shot, as well as their race. We made the following observations:

  1. We see from the boxplot below that the median age for Black people that have been killed by police is 29 years.

  2. White people have a relatively higher median age of 35 years whereas Asian people have the highest median age of around 38 years.

If we look at the age of each victim against the status of their mental health, we can make the following observation: signs of mental illness appear more frequently within the 30s age range while death by police for people age 50 and above are more common for people showing signs of mental illness.

We also looked at the death by race and gender, coming up with the following insight: individuals across all races that were shot and killed by police were more often men.

We then looked at the distribution of deaths by race and the top 5 armed categories. We discovered that around 9% of the Black victims were unarmed whereas only approximately 6% of the White victims were unarmed. Guns were the most used weapon across all races except for Asian individuals. Asian victims were more often wielding knives.

The following graph illustrates the deaths per year by race from 2015-2022:

We looked at the distribution of deaths by suspects’ race and whether they were trying to flee or not. The following are some of our most interesting observations:

  1. Only 53% of Black victims shot were not fleeing whereas 71% of the Asian victims who were shot were not trying to flee.

  2. The car is the most popular method of fleeing among White victims whereas for Black victims, the most popular method of fleeing was by foot.

After performing all of this exploratory data analysis, we decided we wanted to ask a state- and region-based question. Performing a more comparative analysis using this data set could help us understand why police shootings occur in some states over others. This required us to divide the data in a new way. For data visualization purposes, we chose to divide each state into regions. However, our main focus will rely on state-by-state comparative analysis. The following regional subsets include:

Northwest (NW) Southwest (SW) Midwest (MW) Southeast (SE) Northeast (NE)
California New Mexico Illinois Georgia New York
Washington Arizona Wisconsin Alabama Rhode Island
Oregon Texas Indiana Mississippi Maryland
Nevada Oklahoma Michigan Louisiana Vermont
Idaho Hawaii Minnesota Tennessee Pennsylvania
Utah - Missouri North Carolina Maine
Montana - Iowa South Carolina New Hampshire
Colorado - Kansas Florida New Jersey
Wyoming - North Dakota Arkansas Connecticut
Arkansas - South Dakota West Virginia Massachusetts
Arkansas - Nebraska DC -
- - Ohio Virginia -

Fatal shootings in the Northwest United States:

## [1] 1810

Fatal shootings in the Southwest United States:

## [1] 1226

Fatal shootings in the Midwest United States:

## [1] 1080

Fatal shootings in the Southeast United States:

## [1] 1890

Fatal shootings in the Northeast United States:

## [1] 568

After dividing up the data set, we decided to spotlight the body camera variable. This column in the csv file asked whether the body camera was turned on or off during the fatal police shooting. The data was compiled from official police records, completed by the officers on site. In the Washington Post’s brief article introducing the data set and examining some of the most interesting findings, there was no mention of the body camera variable. This provided us an opportunity to create a relevant, meaningful research project on a little-studied aspect of police work. Our final research question is as follows:

Within our data set of 6,574 observations of police shootings from 2015 to 2022 in the United States, is there a correlation between the U.S. state of observation and whether a body camera was turned on during the shooting?

After performing the technical analysis needed to assert that body camera use varies by state, there is still valuable information we need to find a link between state policy on body cameras and the rate at which body cameras are turned on during fatal shootings. The remainder of this paper will attempt to bridge quantitative analysis using the Washington Post data set with qualitative information on state body camera policy. We will also look into funding by state for police work to see if there is a strong correlation between the amount of police funding and the rate at which body cameras are turned on during fatal shootings.

Part I: Technical Analysis

Number of fatal shootings where the body camera was on:

##   body_camera   n
## 1        TRUE 947

Number of fatal shootings where the body camera was off:

##   body_camera    n
## 1       FALSE 5627

We created some figures to visualize body camera use by state and region. In the below bar graph, TRUE signifies a police body camera that was on, while FALSE indicates the body camera was off:

A very important variable allowed us to study body camera use variation by state. This variable is stbcp, or the proportion of shootings when a police body camera was turned on by state. We first checked our data for normality. Because the plot below is relatively linear, we concluded this data is close enough to normality for our purposes.

This scatter plot below illustrates the comparative power of stbcp. Each point on the graph depicts a state’s proportion of shootings where the police body camera was turned on during the incident. These data points also show an interesting regional variation. For example, we can see that there is very little variation in Southwest, and many differences among states in the Midwest.

We performed a chi-square test to understand whether there is a significant difference between the proportions of body cameras being on during fatal shootings each state.

\(H_{0}\): There is no significant differences between US States in the proportion of body cameras being turned on during police shootings

\(H_{A}\): There is a significant difference between US State in the proportion of body cameras being turned on during police shootings

Significance Level: \(\alpha = 0.05"\)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   0.000   0.101   0.133   0.144   0.183   0.409
## 
##  Pearson's Chi-squared test
## 
## data:  contable
## X-squared = 3e+05, df = 2300, p-value <2e-16

With a p-value of less 2e-16, we easily pass our significance level of alpha=0.05 and have shown that there exists significant differences between different states’ proportions of body camera usage during fatal police shootings. The challenge now becomes, what is causing this variation in state use of body cameras during active police work? We intend to delve into the reasons why there are differences and research what factors may explain these differences between states. This will require understanding state laws and policies regarding the use of police body cameras. We must also understand the police force consequences for turning off body cameras during police activity in different states.

Studying the use of body cameras in police work is an important topic of study for data-driven policy research in the United States. We hope to be able to apply this correlation between the U.S. state of observation and whether the body camera was on or off during the shooting to state policy on body cameras during police work.

Part II: Results

We have the visualization, as well as the Chi-square test, to confirm that there is a significant difference in stbcp (the proportion of shootings when a police body camera was turned on by state). Let us now look at the rankings by state of stbcp, from most to least likely of having a body camera on during a police shooting.

State stbcp State stbcp State stbcp
DC 0.409 Vermont 0.333 Nevada 0.330
Utah 0.307 Maryland 0.277 Connecticut 0.238
Iowa 0.225 Hawaii 0.212 Nebraska 0.200
Rhode Island 0.200 California 0.188 Idaho 0.185
Indiana 0.185 Oklahoma 0.179 Arizona 0.161
Alabama 0.160 New York 0.160 New Mexico 0.159
South Carolina 0.156 New Hampshire 0.150 Illinois 0.148
Wisconsin 0.146 Virginia 0.144 North Dakota 0.143
North Carolina 0.134 Michigan 0.133 Minnesota 0.133
Ohio 0.131 Texas 0.131 Kansas 0.130
Delaware 0.125 Colorado 0.123 Louisiana 0.119
Washington 0.103 Kentucky 0.103 Tennessee 0.101
Montana 0.100 New Jersey 0.100 Florida 0.0995
Georgia 0.0988 Alaska 0.0952 Arkansas 0.0787
Oregon 0.069 West Virginia 0.0678 South Dakota 0.0625
Missouri 0.037 Pennsylvania 0.0352 Maine 0.000
Wyoming 0.000

Here we can see that the states with the highest number of observations where the police body camera was turned on include DC, Vermont, and Nevada. The states with the lowest number include Pennsylvania, Maine, and Wyoming–Maine and Wyoming having no recorded police shooting where a body camera was turned on. Let us now do a deep-dive into state policy on body cameras for these six states.

District of Columbia

Vermont

Nevada

Pennsylvania

Maine

Wyoming